Development and Validation of Novel Optimization Approaches to Solve the Flexible Job-Shop...

Registro bibliográfico

  • Título: Development and Validation of Novel Optimization Approaches to Solve the Flexible Job-Shop Scheduling Problem = Desarrollo y validación de nuevos enfoques de optimización para solucionar el problema de ordenamiento de tipo Job-shop flexible
  • Autor: García León, Andrés Alberto
  • Publicación original: 2016
  • Descripción física: PDF
  • Nota general:
    • The classical job-shop problem is among the most studied scheduling models because it can model a variety of real-life applications. This problem can be reinforced by integrating the operation flexibility to obtain a more realistic model, called Flexible-Job shop Scheduling Problem (FJSP), that capture the salient features of modern systems.

      To consider both productivity and customer service in the FJSP, objective function such as regular criteria that consider an increasing function of the completion times of the jobs are needed. Makespan consists of minimizing the maximal completion time of all jobs, and it is the most studied criterion in the FJSP. However, minimizing other regular criteria are better suited to capture critical factors that affect the profit of manufacturing and service systems and thus their competitiveness; in particular the importance of jobs(customers) with weighted criteria.

      This thesis presents innovative general local search approaches to minimize regular criteria in the FJSP in two ways: Single criterion and multi-criteria called Multi-Objective Flexible Job-shop Scheduling Problem (MOFJSP). The local search approaches make use of a generalization of the disjunctive graph model, iteratively progress from one solution to another using two neighborhood structures (N1 and N2). N1 consists of moving all critical operations that belong to critical paths of jobs which affect the criterion being optimized and N2 is a subset of N1 where only operations which belong to blocks. New, efficient and fast sufficient feasibility conditions to obtain moves of the neighborhoods and evaluation functions that assess the quality of the feasible moves are proposed.
  • Notas de reproducción original: Digitalización realizada por la Biblioteca Virtual del Banco de la República (Colombia)
  • Notas:
    • Resumen: Búsqueda local y optimización Mono-objetiva; Criterios regulares; Job-shop flexible; Local search; Mono-objective optimization; Optimización Paretiana; Pareto optimization; Regular criteria
    • © Derechos reservados del autor
    • Colfuturo
  • Forma/género: tesis
  • Idioma: inglés
  • Institución origen: Biblioteca Virtual del Banco de la República
  • Encabezamiento de materia:

Web semántica